Размещено на http://www.allbest.ru/

GOVERNMENT OF THE RUSSIAN FEDERATION

NATIONAL RESEARCH UNIVERSITY

HIGHER SCHOOL OF ECONOMICS

FACULTY OF WORLD ECONOMY AND INTERNATIONAL AFFAIRS

MASTER OF INTERNATIONAL BUSINESS PROGRAM

MASTER THESIS

Topic: “Factors Driving Business Model Adaptation to Multi-sided Platform Reality by Global Retail Industry: Case of Amazon”

Student:

Ekaterina Logina

Moscow 2020

Content

Introduction

Chapter 1. Theoretical aspects of multisided platforms (MSPs) and a literature review on the topic

1.1 The concept of the multisided platform: an overview of definitions and the evolution of approaches

1.2 Multisided platforms as a new business model

1.3 Factors driving business model adaptation to the multisided platform business model approach in retail industry

Chapter 2. Analytical part

2.1 Retail industry outlook: trends and relevance of participation in digital platforms

2.2 Case study of Amazon company: presence of factors driving traditional BM adaptation to MSP BM

Chapter 3. Outcomes and recommendations

3.1 Key findings on factors driving business model adaptation to a new MSP BM in retail industry

3.2 Limitations and further research on the topic

Conclusion

Bibliography

Introduction

The modern business environment around us is in a constant change nowadays. On the one hand, there appear more and more companies from large corporations to small family businesses and startups every day, and each of them competes for and tries to take a bigger stake in revenues by getting more clients by investigating all their needs, interest, consumer journey, barriers for purchasing . On the other hand, consumers are aware of the presence of a large variety of companies and products on the market, so, customers have become very demanding about products they buy: people nowadays try to obtain all available information about the product they buy, compare its characteristics with one product substitutes have, read as many reviews and feedbacks as possible before actually making the purchase. All of the above changes in the business environment have become present due to the information technologies that transform economic conditions and lead to the appearance of new business models. One of the most significant changes in the economic and business sphere of recent years is the development of MSPs - multisided platforms. Multisided business platforms have been existing for a very long time in some form, but as we know them today, mostly as digital platforms, they have started their rapid development with the spread of the Internet. Companies using a multisided platform business model are the most recognizable and often-used among consumers, and the most rapidly growing and prosperous in terms of earning money among all other companies. Such companies as Uber (taxi), Airbnb or Booking (hosting and traveling), Amazon or Alibaba (retail), Google (search & information), Facebook (communication), Visa (payment system) are only a few examples of very successful companies based on a multisided business model.

Development of the multisided platform as a new way to arrange the business model of a company affects lots of spheres of business, and the retail industry is among them. Who needs to go to an actual retail store, spending valuable time on the road, when there exist such retail giants as Amazon or Alibaba, who enables consumers to choose all products they need by comfortably sitting on their sofa at home and get those products delivered within several days. Many researchers dedicate their works to an investigation of how such platforms as Amazon, Uber o Facebook work, how much they earn, or the process of transition from traditional business models to ones that include platforms. But there is a lack of knowledge of why companies transform their existing model to a platform one, why should they do so, or what factors overweigh and motivate them to do so, especially in the retail industry.

So, the goal of this master thesis is to reveal factors driving retail companies' participation in the multisided platform business model. To achieve this goal, the following tasks were set:

· Conduct an existing literature review on the topic, which include: an overview of multisided platforms concept, key definitions, the evolution of approaches and development of the concept;

· Differentiate features of MSP as a new business model

· Compare MSP BM with traditional BM, which includes linear company's operations

· Analyze MSP BM relevance to retail industry companies business activities

· Conduct analysis of international retail companies participation in MSP BM by case study analysis

· Reveal factors facilitating international retail companies participation in MSP BM and further perspectives for companies' participation in MSP BM

Therefore, the object of this master thesis could be defined as the multisided platform as a new business model, while the subject of the research is factors facilitating BM adaptation to the MSP by global retailers. As a result the research questions could be formulated as follows: What factors drive retail companies participation in the multisided platform business model? Which of these factors are present for the Amazon company? The research hypotheses could be considered as follows - for each factor revealed and analyzed in the theoretical part, we could formulate hypotheses in the analytical part as:

H₀: “i” factor is present for the Amazon company.

H₁: “i” factor is not present for the Amazon company.

Theoretical bases for the master thesis include three main sources. Firstly, lots of attention was paid to the papers of scholars and writers, who specifiy their attention to the topic of multisided platforms, such as (Parker, Van Alstyne, Yablonsky, and others). Also, well-recognizable reliable journals were applied to conduct research, for example, Harvard Business Review and Forbes. Moreover, different analytical reports and databases such as Statista, Euromonitor, Price Waterhouse Coopers, Deloitte, and Similar Web were very useful and helped to make the research more representative.

The concept of platform business model itself is not new if we consider shopping malls or fairs. But multisided markets as digital platforms mostly present today are quite a debating and undisclosed topic in many ways. Digital platforms have gotten their huge boom with the appearance of the internet. And their development has become actual only with the spread and use of the internet by the majority of the world population. Existing literature on the topic of MSP BMs mainly considers the development and ambiguity of definitions, investigation of the model itself, or process of transition from a traditional business model to a platform one. But there is a lack of scientific research on topics of why traditional companies transfer from traditional to platform BM or should do so or what factors motivate companies for this transition.

This master thesis consists of three chapters. The first chapter pays attention to the existing literature review on multisided platforms: existing definitions and concepts; theoretical approaches and classifications; and how MSP approaches to be a new business model, determining its distinctive features as a new business model and its differences from traditional linear business model and MSPs development in the retail industry. In the second chapter, we consider the relevance of retail companies' participation in the MSP business model and by what economic changes and trends this relevance could be explained. After that, the case study of Amazon (one of the hugest digital retailers in the world) operations was conducted to show how retail companies manage to implement MSP as their business model and how that differs from the retail companies conducting their operations in a traditional linear way. The third chapter is devoted to the summary of MSPs as a new business model phenomenon, its development in and effects on the retail industry and main factors driving adaptation from a traditional BM to a MSP BM. It also includes some recommendations for further companies participation in MSP BM and limitations and spots for further research of this work.

business amazon retail trade

Chapter 1. Theoretical aspects of multisided platforms (MSPs) and a literature review on the topic

1.1 The concept of the multisided platform: an overview of definitions and the evolution of approaches

Every step of our life in the modern world is surrounded by platforms. Want to get in touch with your friends, you go to use Facebook or Instagram; want to go somewhere by car, you use Uber to get your taxi, want to go traveling, you use Booking or Airbnb, the same story with ordering goods by using Amazon or Alibaba. And there are many more, like Netflix, Google, etc., so consumers barely need to go away from home to get their needs to be met. Platforms are on their rise and surround us everywhere nowadays. So, this chapter is devoted to theoretical aspects of multisided platforms, an overview of definitions, concepts, existing literature on the topic and the evolution of approaches.

Even though multisided platforms have been researched for around twenty years now, it is still quite a new and debating notion and there is no clear and widely accepted definition for them yet (OECD 2009). According to the publications on platform economy dynamics on Scopus, in the period from 1991 to 1999, there were almost no publications on this topic. But after that, the dynamics of platform economy publications on the Scopus has started to increase gradually and accounted for around 200 units in the year 2010. While for the year 2017, that number has raised significantly and accounted for around 700 publications. (Scopus, 2019). One of the first definitions for MSP was proposed by Rochet and Tirole (2003), where platforms and markets were considered the same thing: “A market with network externalities is a two-sided market if platforms can effectively cross-subsidize between different categories of end-users that are parties to a transaction.” The great contribution of this definition is the implementation of the difference between one and two-sided markets, that implies that what really matters is the fact who pays for the service. For example, taking credit cards' markets, it is the sellers who pay for the transactions, not the buyers. Because otherwise, or even in the case were buyers and sellers pay equally, there would be much fewer transactions as fewer buyers will use credit cards, and, as a result, fewer sellers will offer such a payment. But there are also some drawbacks in this definition for which it was arguable by other authors: 1) price structure is the only factor to consider, 2) implies the fact of control of all transactions by the platform. For example, there are other types of platforms, such as newspapers with advertising. And platforms in this case usually cannot control whether readers are interested in advertising or not (Exemplified by Saмnchez-Cartas J., M., Leoмn G., 2019 ). Then, there were introduced several more broad definitions. “Multi-sided platforms coordinate the demand of distinct groups of customers who need each other in some way” Evans (2003). That definition does not imply control of every transaction by a platform, but it was too broad. Or, the presence of indirect network effects to be vital for defining a multi-sided platform: “the net utility on the side “i” increases with the number of members on side “j” ”, also introduced by Rochet and Tirole, but also criticized by other authors as too broad. And finally, one of the most recent and full definitions was proposed by Weyl (2010), where the multisided platform is defined as market obtaining the following features:

· There is a multi-product firm. A platform provide distinct services to two/or more sides of the market

· There are cross networks effects. Users' benefits from participation depend on participation on the other side of the market

· Bilateral market power. Platforms are price setters on both sides of the market

Weyl emphasizes that there could be some drawbacks in this definition, but the absence of any of the three features mentioned above makes it simpler and better to define such a case as standard or one-sided market.

This definition is true for all MSP we know. Let us consider how the definition works by the example of one of the most successful and well-known MSP present on the market : 1) Airbnb is the online platform that matches tenants and landlords from all over the world. For tenants, it is a place where they can find housing for a living or a vacation, for landlords it is a place to find potential tenants and earn on renting housing they own - 2 distinct services. 2)The more landlords are present on platform, the higher the benefit for tenants as there is higher completions and better prices and a choice is wider. It also works in an opposite direction. 3) Airbnb as a platform is a price setter for both landlords and tenants, and the company requires a fee from every transaction, which is around 3% for a landlord and around 14% for a tenant. (Airbnb Official Website).

And, as it is mentioned by Filistrucchi et al. (2012a): “although at first sight there are appears to be still some debate on the exact definition of a two-sided market, the different definitions proposed appear consistent enough to allow the practical identification of two-sided markets.”

What is more interest for us to understand, is the modern definition of MSP, or definition of so called online or digital platforms. The definition proposed by OECD in 2019 or by EC in 2015 for online platform is “intermediary service provider, undertaking operations in two (or multi) - sided markets and using the Internet to enable interactions between two or more distinct but independent groups of users to generate value to these groups”. Precisely, the wide propagation of the internet in peoples' lives makes MSP so popular and successful form of business model recently. Most of more prosperous and well-known multisided platforms are online platforms, because the Internet makes it much easier: 1) to escape the need to be “there” in the location of marketplace to use its advantages, 2) give higher potential for spread among users, and, as a result, higher cross network effect, 3) helps to eliminate costs of different kinds (transportation, spread, search, etc.).

Scientific foundation of literature on multi-sided platforms is twofold. Some people refer to the first paper considering indirect network effects (Parker and Van Alstyne 2000) as a birth of multi-sided literature. Others, in their turn, believe that scientific foundation was born with the first paper considering “economic platform” (Caillaud and Jullien 2001; Rocket and Tirole 2003). Nonetheless, the very first scientific research which describes the specific characteristics of multi-sided markets is Wright (2004). In his work, the author examines eight general statements in the literature of industrial organizations that are NOT relevant to multi-sided markets. So, going from the opposite, those eight fallacies are as follows:

* An efficient price structure should be set to reflect relative costs (user-pays)

* A high price-cost margin indicates market power

* A price below marginal cost indicates predation

* An increase in competition necessarily results in a more efficient structure of prices

* An increase in competition necessarily results in a more balanced price structure

* In mature markets, price structures that do not reflect costs are no longer justified

* Where one side of a two-sided market receives services below marginal cost, it must be receiving a cross-subsidy from users on the other side

* Regulating prices set by a platform in a two-sided market is competitively neutral

As it is argued by Evans and Schmalensee (2013), “one-side tools often do not apply, at least not without substantial change, to multisided platforms ”. The main reason for that is the fact of interrelationship between sides present on multisided markets. This reason was raise in many MSP researches, and all of them came to the conclusion that most of insights applicable to traditional markets are not such in relation to multisided markets.

As a general rule, as it usually happens when something new appears, literature on the subject of multisided platforms developed through a comparative analysis of multisided platforms and traditional unilateral markets. Let us look at the main similarities and differences emphasized in the works of scientists having weight in the gross of multilateral markets in the form of the following comparative table:

Table 1: Overview of Main Similarities & Differences (multi-sided markets VS one-sided markets) in Existing Literature on The Topic.

Differences
Multi-sided markets	One-sided markets
Pricing
Tend to set asymmetric price structure in which one side- profitable, other-loss side. (Caillaud and Jullien (2001), Caillaud and Jullien (2003) or Schiff (2003))	An efficient price structure should be set to reflect relative costs (user-pays). (Wright (2004))
The prices can be totally disconnected from marginal costs. (Evans (2003))	A price below marginal cost indicates predation. (Wright (2004)) A high price-cost margin indicates market power. (Wright (2004)) In mature markets, price structures that do not reflect costs are no longer justified. (Wright (2004))
Interrelationship between the sides make it possible for a platform to respond to an increase in costs on one side with an increase in prices on the other side. It is common in multi-sided markets that the welfare-maximizing prices will not coincide with marginal costs. (Jullien (2005))	Where one side of a two-sided market receives services below marginal cost, it must be receiving a cross-subsidy from users on the other side. (Wright (2004)) Regulating prices set by a platform in a two-sided market is competitively neutral. (Wright (2004))
It is hard to establish whether a given price in a two-sided market is higher or lower than socially optimal, or even whether greater competition would make the existing price rise or fall. (Rysman (2009))	An increase in competition necessarily results in a more balanced price structure. (Wright (2004)) An increase in competition necessarily results in a more efficient structure of prices. (Wright (2004))
Similarities
Multi-sided markets	One-sided markets
Network economics
Pessimistic beliefs are very difficult to overcome. When one side believes no one is going to enter the platform ( pessimistic belief), it can get a positive market share if a platform adopts a divide-and-conquer (DC) strategy - the platform must subsidize one group to attract the other group. (Caillaud and Jullien (2003))
Differentiation
The greater is the differentiation, the greater are profits, at least for one platform/company. (Reisinnger (2004)) If platforms/ companies can differentiate from each other, they can coexist in the same market. (Rysman (2009))
One indirect network effect away
Two-sided market model leads to the internationalization of externalities, to bigger profits, and higher welfare levels. However this result is only true if externalities exist, if not, there is no difference between two cases. (Parker and Van Alstyne (2005))

Source: made by the author of the research based on the work of Saмnchez-Cartas J., M.,, Leoмn G., 2019.

Despite the fact, that scientific research on the topic related to multi-sided platforms has appeared around twenty years ago, the phenomenon of “economic platform” as we used to know it (as a new way of business model for the majority of most successful digital companies nowadays), starts to develop with the boom of the internet and social networks, as it enables companies to build their business models via online technologies. It has led to a noticeable increase in the development of the concept of multi-sided digital platforms itself; in number of firms, using such a business model; and, in the number of scientific researches related to the topic. Hence, as scientific researches that devoted directly and explicitly to MSP is a quite recent phenomenon, it could be concluded that research related to this topic is just in its beginning.

Digital revolution has played a huge role in the development of platforms -starting from the point that it makes the range of opportunities for businesses unlimited to the fact that it changes to the very core all business activities and the way they are performed. In the Table 2 below you can see examples of most prosperous in their sectors existing companies that use MSP as their business model:

Table 2: Examples of Companies Using MSP by Different Activities.

Sector	Examples of companies	For whom:
Accommodation & tourism:	Airbnb, Booking	Tenants, landlords, tourists, advertisers, hotels
Transportation: airplanes and train	Onetwotrip, Skyscanner,	Tourists, advertisers, avia companies
Transportation: carsharing, taxi	Uber, Lyfth, Belkacar, Youdrive, Citymobile	Drivers and those who need a ride
Learning platforms:	Coursera, Lingualeo, Gickbrains	Teachers and students
Food delivery:	Deliveryclub, Yandexmarket	Restaurants and consumers
Entertainment:	Netflix, Youtube, Apple music, Spotify, Vine	Content makers, bloggers, advertisers, marketers and content producers
Career related:	Headhunter, LinkedIn	Employers, recruiters and employees
Social networking:	Facebook, Instagram, VK, OK, Wechat	Social network users, advertisers, marketers, content makers, content producers, bloggers
Dating related:	Tinder, Badoo	App developers, advertisers, Date app users
Search engines:	Google, Mail, Yandex	Search engine users, advertisers, app developers,
App related:	App store, Android store	App users and developers
Payment systems:	Visa, PayPal, Apple pay	Buyers and sellers, individuals and businesses
Retail:	Alibaba, Amazon, eBay	Sellers, retailers, buyers, advertisers
Games related:	Xbox, Sony Playstation	Game developers and game players

Source: made by the author based on literature review and market analyses.

As it could be concluded from the table above, most of the services we use every day that make our life so easy, actually are multi-sided platform, while several decades ago it was traditional linear business model companies. That fact actually makes this topic even more relevant and interesting for the investigation.

On the one hand, MSPs is a topic of a current interests. On the other hand, this topic is very debating and is not fully covered by recent scientific researchers. As a result, sometimes it is confusing, whether market or platform is multisided o two-sided, or even it is simply a reseller. To make it more clear, let us discuss main characteristics, which define MSPs.

According to Yablonsky (2013), A multi-sided market exists if at any given time the following factors are satisfied:

· The presence of two or more different sides (groups) of participants (users);

· Value obtained on one side, increasing with the number of participants on the other side (on other sides);

· The presence of an intermediary (platform) to ensure direct interaction of participants of various parties and to ensure network effects created on one side for the other side;

· Participants are permanent members of one separate group (party), which carries out transactions with participants in the second group or several groups. It should be noted that in unilateral markets there are no stable permanent groups of participants [Wright, 2004].

· User participation in one of the parties to the MSP should be accompanied by a certain degree of affiliation with the platform. Awareness of the decision can be expressed in a willingness to incur costs (for example, registration on a site, etc.), or willingness to pay an entrance fee (purchase of a video set-top box or membership on a dating site). This requirement is aimed at eliminating the erroneous assignment of suppliers of trading floors, equipment or services that are not related to business processes that occur within the MSPs.

· In order to distinguish MSPs from the resellers or organizations buying goods or services and then reselling them to the final consumer, it was proposed to refer to MSPs only those organizations that provide users with the opportunity to directly interact with each other ( communication, exchange, consumption, or a combination thereof), while retaining control over the basic conditions for joint activities of user-consumers.

· So called chicken and egg dilemma can be overcome by due to the fact that different sided charged differently: one side subsidizes a participation of the other side.

Picture 1: The Difference Between Multilateral Platforms, Resellers, and Retailers.

Source: Hagiu, Wright, 2011.

So, it could be concluded, that the topic of multisided platforms is quite a relevant but debating research area. It has different definitions, components, theories of evolution and classifications and lots of area for further research. In this work, we would apply to the definition, components and classification given by Sergey Yablonsky, mentioned above. As his work is one of the most recent researches that more fully explained the concept and considered all previous most influential researches on the topic.

1.2 Multisided platforms as a new business model

Before talking about multisided platforms as a new business model, let's give a definition to the concept of business model itself and have a look at evolution of the concept for a better understanding of the topic. The concept of a business model starts its development close to definition of strategy given by Michael Porter in 1980, which is “deliberately choosing a different set of activities to deliver a unique mix of value”. Peter Drucker introduced understanding of business model as a set of assumptions about what business will and will not do, about what company is paid for, about markets, customers, competitors, their values and behavior, about strengths and weaknesses, about technology and its dynamics in his Harvard Business Review article in 1994.

After that the evolution of the concept refers to Michael Lewis. In 1999, he defines business model as “All it really meant was how you planned to make money” to explain the dot.com bubble in the simplest way. In his later work of 2014 “The New, New Thing” the author defines business model as “a term of art”.

After that, Joan Magretta in her work “Why Business Models Matters” in 2002, as a corrective to Lewis, cites Ducker in definition of business model. She personally defines business model as “at heart stories that explain how enterprises work” and believes that a good business model should answer Peter Ducker's age-old questions: “Who is the customer? What is the customer value?” and fundamental questions all good managers ask themselves: “How do we make money in this business? What is the underlying economic logic that explains how we can deliver value to customers at an appropriate cost?”. Joan Magretta, as well as Peter Drucker, is focused more on assumptions that business should put forward rather than on money. This focus is based on the information that the concept of business model was developed with the appearance of personal computers and spreadsheets, which make it possible to test and model various components of business. “Before that, successful business models were created more by accident than by design or forecast, and became clear only after the fact. By enabling companies to tie their marketplace insights much more tightly to the resulting economics - to link their assumptions about how people would behave to the numbers of a pro forma P&L - spreadsheets made it possible to model businesses before they were launched” said Joan Magretta. Since her view of business model is in business modeling, the author further defines a business model in terms of value chain, which has two parts: first part includes everything associated with product creation (design, raw materials, manufacturing, etc.); second part includes everything associated with selling products/services (target customers, transactions, distribution, delivery, etc.).

One of the most well-known and recent authors related to the concept of business model is Alexander Osterwalder, who also see a concept of business model as a set of assumptions. The author has developed nine-part business model canvas (2005-2008), which enables to construct those hypotheses about one's business model. The canvas includes nine most important assumption about a business to understand what the business is and how it works. They are about: key resources, key activities, key partners, value propositions, customer relationship, customer segments, channels, cost structure and revenue streams.

The model and its key components could be better considered by the example. And consideration of a traditional retail company such as Walmart would be relevant for our topic. The Walmart is the biggest offline retail company in the world. Its main business activity involves physical retail stores. Although, the company tries to incorporate an online presence to its strategy due to digital market trends.

Consideration of the Walmart business model on the Osterwalder Business Model Canvas includes:

Table 3:Walmart Business Model Canvas.

Key Partners:	Direct suppliers, landlords, investors, manufacturers, international labor organizations.
Key Activities:	Inventory management systems, vendor control negotiations, supply chain management approach, shrink management.
Key Resources:	Vendor relationship, distribution system, highly skilled management, purchasing & procurement network (a new step to digitalization).
Value Proposition:	“Offering everyday lowest price on a wide variety of basic products”.(Walmart, 2020).
Customer Relationships:	Mobile apps (a new step to digitalization), advertisement, mailing (a new step to digitalization).
Customer Segments:	Low income, frugal, rural customer base.
Channels:	Physical stores (the main channel), online store (a new step to digitalization).
Cost Structure:	Payrolls & benefits, rent, costs of goods sold, logistics, maintenance, utilities.
Revenue Streams:	Financial services (a new step to digitalization), commissions, retail sales, warranties & insurance (a new step to digitalization).

Source: made by the author based on the Walmart official website information.

As different business models usually describe different kinds of businesses, A. Osterwalder and Y. Pigneur also have developed several patterns of business models. Their classifications consists of five different patterns:

1. Unbundling business models

This pattern of business model includes the fact that there are three fundamentally different types of business within one corporation: customer relationship business, product innovation business and infrastructure business. They are fundamentally different because their economic/competitive/cultural plights are different. They can coexist within one corporation but in fact they are so called “unbundled” into separate entities in order to avoid conflicts. (examples: mobile telecom industry/private banking industry)

2. The long tail

This type of pattern focuses on selling less of more. It is mainly about a niche product, which you offer in a large number but sell quite infrequently. In this pattern a small number of bestsellers makes most revenues, but it requires low inventory costs and strong platforms to make niche products ready and available to interested customers. (examples: Netflix, Facebook, etc.)

3. Multi-sided platforms

Pattern of business model that connect two or more distinct groups of customers. The platform creates value by providing interactions between different groups. And value for a group of customers is created only if the other group is present. What is more, the more users the platform attracts, the more value it creates for the customers (network effect) and the more it grows. (examples: eBay, Google, etc.)

4. FREE as a business model

In this type of business model pattern, at least one segment of customers is able to benefit from a free-of-charge offer. It is possible because of two things: 1- different patterns are used, 2-customers who do not pay are financed by other segments of customers who pay (usually it works like that: customer gets basic services for free and more comprehensive services are paid). (examples: Skype, Google)

5. Open business model

Used for a possibility to make systematic collaborations with external partners and gain value from it. This pattern may be used to exploit external ideas or to provide internal ideas as an asset for external parties. (examples: P&G, GlaxoSmithKilne, etc.)

As it could be seen from the business models classification, MSP is one of them. Platforms have existed for years, for example, malls connecting merchants and consumers also were kind of platforms. But what happens in this century is the fact that due to the appearance of the Internet and different wireless technologies, the possibility to create new organizational forms of business appears. (Eckhardt et al., 2018; Iansiti and Lakhani, 2017; McIntyre and Srinivasan, 2017; Parker and Van Alstyne, 2018; Teece, 2018). And, as you can see from the table 2 in the paragraph 1.1, nowadays companies whose business model is the multisided platform are present almost in all spheres and industries, so they are very different. As companies are from different industries and serve different goals (services/products/users differ), it is logical that platforms are also differ from each other, but there are several parts of ecosystem that always remains the same, and make the platform itself. These parts are usually attributed to the platform structure and players:

· Platform owners

· Providers

· Producers

· Consumers

Picture 2: Platform Business Model.

Source: Parker, Van Alstyne, HBR 2016

Platforms owners control their intellectual property and governance. Providers create interfaces for the platform used by different sides. Producers create products or services offered to customers. And consumers use this offerings.

The next step of our research is to understand how companies move from the traditional, so called “pipe-line” business model to the multisided platform business model. Pipeline business model makes products or services through the classical value-chain model value creation. According to Porter (1985), inputs at the beginning of the chain (e.g. raw materials) should go through a series of steps to be transformed and bring some additional value to consumers at the end of the process as a finished product. Those steps include inbound logistics, operations, outbound logistics, marketing and sales and services as primary activities and procurement, human recourses management, technological development and infrastructure as a secondary activities. As a result, the chain of activities altogether creates higher additional value for the product than the sum of added values of all activities separately.

Picture 3: Michael Porter's Value Chain.

Source: Michael Porter, 1985.

For example, Apple viewed by its hardware could be a good example of the traditional pipeline model. But Apple's software- App Store is a platform business model that aims to connect apps developers and their users. What is more, by this example we can see that companies can do more - they can combine both business models.

There are lots of firms out there, that still pretty successfully operate as a classical pipe business models and earn huge money. But platforms that were successful in entering the market, usually outperform existing traditional players. This is one of the vital reasons lots of traditional value-chain-model companies try to incorporate platforms in their business models. According to Van Alstyne, Parker and Choudary (2016), a transformation from the traditional model to the platform business model includes the following steps:

· Change of the focus from the control of resources to the orchestration of resources: in the traditional model, a firm is more competitive than others, if it owns more scarce and valuable resources such as real estate as tangible assets or intellectual property and goodwill as intangible assets. In the reality of platforms, the valuable and scarce recourse is a platform community with its network and resources that users bring with them (it could be users' cars, houses or even ideas).

· Change of the focus from the internal optimization to the external optimization: as the classical business model creates value through the value chain, it could increase value of the finished product for the consumers by optimizing a value chain processes in its every step. Platforms, on the other side, create value by creating ecosystem, where the main value is a network between users. So, it is a connection of external producers and consumers. That usually leads to the reduction of even variable costs to zero. As a result, platforms do not need to control the process of product creation, they need to optimize the process of ecosystem governance and to attract more users to the platform to increase value for clients- optimize external processes.

· Change of the focus from the customer value to the ecosystem value: traditional business models aim to maximize the lifetime value of individual customers of products or services, while platforms aim to maximize the value of the whole ecosystem in an iterative, feedback driven, circular process. Sometimes, it includes subsidies for participation of one group of users in order to attract another group of users.

· Change of the units of analyses from the goods/services delivered to interactions: for the traditional pipelines companies managers usually focus their attention on revenues and profits, or, more precisely, on an increase of these indicators. So, goods or services of a company serve as units of analyses. In an economy of platforms, as it was already mentioned above, companies focus more on interactions between groups of their users. Platforms try to achieve higher number of interactions rather than higher sales, as its associated with higher network effect and gain more value for companies. That is why it is critical for platforms to create right user-friendly design that will not only attract users, but also enables higher number of interactions between them.

· Change of the metrics focus: quite a narrow scope of metrics was considered as a scope for evaluation of companies health and performance success. Managers in traditional pipeline companies usually consider such ratios as inventory turnover, process optimization, opening bottleneck, margins, rates of returns and some other standard metrics. But as focuses of companies with the introduction of platform have changed, metrics that should be monitored also have changed. Some of new metrics, that are crucial to track for successful platform performance are as follows:

1. Match failure: if a user opens an AppStore and do not find an app he or she is looking for, the match is failed. And when this user fails to find an App of an interest several times, this user will stop using the AppStore and probably will use something else. Match failure decreases a network effect.

2. Engagement monitoring: healthy platforms should monitor users' participation that increase network effect. For example, social media, such as Instagram or Facebook, track periodic users' participation to evaluate their activities dedicated to enhance engagement.

3. Interaction quality: platform should monitor not only the success or failure of interaction or number of interactions, but also their quality. It is important whether users were successful in finding exactly what they wanted.

4. Network effect: most of the time a network effect is considered as a positive externality, but sometimes it could be bad. When the platform is badly managed and users' misbehavior is present, it could lead to a negative feedback loop.

5. Financial value: considering financial value of the platform, metrics beyond traditional financials should be considered. For example, values of platform's community and network effect.

· Change of the focus from the supply-side economies to the demand-side economies: another very important thing that should be emphasized while talking about multisided platform as a new business model is a positive externality, appearing in MSP BMs and not so present in traditional ones - the network effect. Traditional model companies are representatives of industrial economy, which main engine is economies of scale from the supply-side. Such companies usually have quite high fixed costs and relatively low variable costs. Higher volumes of sales allows companies to lower their fixed costs, hence, decrease prices, increase volumes of sales further and again decrease fixed costs -feedback loop that produces monopolies. (GE, Rockefeller's Standard Oil) So, such companies get their market power by controlling resources, increasing their efficiency and preventing competition. In contrast, the driving force of internet platforms is a demand side of the economy, or, more precisely, demand-side economies of scale known as a network effect. It is, moreover, enhanced by technologies, which increase efficiency of social media, demand aggregation, applications development and other phenomena that expand networks effect. In the internet economy, the driving force for companies is also volume, but it is the volume of platform participants. The higher is the number of participants, the higher is the price of each transaction. The reason for that is the higher match between supply and demand and the greater amount of information to find this match provided by a wider network effect. A bigger scale creates bigger value and attracts greater number of participants - another virtuous feedback loop that enables monopolies existence. (Alibaba - 75% of all Chinese e-commerce transactions, Google - 82% of mobile operating systems).

The network effect is interesting to consider not only from the side of model focus transformation, but also from the side of considering Five Forces model of Porter differently. Porter's Five Forces analysis (1979) is a grounding model for business competition analysis from everlasting.

Picture 4: A Graphical Representation of Porter's Five Forces.

Source: Porter, HBR, 1979.

According to Michael Porter and his work of 1979, in a competition for a market share, companies usually face rivalry not only in its most trivial form as other players, but rather in a wider form of industry competition as competitive forces which usually include customers, suppliers, potential entrants and substitute products. All of these components are rooted in underlying industry economics and could be more or less prominent and active competitors that should be considered. Usually, they are considered as Porter's basic five forces and could be seen in the Picture four above. The collective accumulation of this five forces usually determines a potential profit in the industry. According to the author, there exists a range for this profit according to collective strength of five external forces. For example, industries like steel have an intense potential profit and huge returns to investments, while industries like soft drinks have a mild potential profit. The main point of the author's work is the fact that whatever the collective strength of forces in the industry is, the strategy of the company is to find such a position in the industry where the company could better defend itself from these forces. Let us look on rivalry forces closer:

· Threat of entry: new entrants are external negative force for the companies because they want to gain some market share on the market that could lead to decrease in profits of existing players. That is why companies usually try to build strong barriers to entry for newcomers, such as economies of scale, product differentiation, capital requirements, cost disadvantages (acquisition of technologies, learning curve, etc.), access to distribution channels, government policy.

· Bargaining power of buyers & suppliers: high bargaining power of suppliers is usually a disadvantage for companies as they could increase prices of raw materials which would decrease quantity of sales or disable companies to raise prices for products and lower companies' profits. On the other hand, high bargaining power of buyers implies that consumers of companies' products could raise their requirements of quality or force prices down which also leads to lower profits for companies in the industry.

· Substitutes: presence of substitutes is bad because of two things: 1) it puts boundaries on a price range of products, 2) it decreases industry potential and also lowers profits as a result.

· Existing competitors: existing competitors are companies that already exist in the same industry and split profits of the industry among each other. Intense rivalry in the industry is usually expressed by price wars, new product introductions, or advertising. What is more, as an industry matures, its growth rate changes. That usually leads to decrease in profits and shakeouts. So, it quite important to know existing industry players to create a strategy of competition with them and protection against their rivalry actions.

Knowing the information about competing forces in the industry allow companies to evaluate its strong and weak points and create so called SWOT analyses (strengths, weaknesses, opportunities and threats). Company's SWOT, in its turn, make it possible to protect a company from rivalry competing forces in the best way, as it was already mentioned.

And here we are back to network effect and why it is interesting to consider it within five forces model. Porter's five forces model does not consider a network effect or the value it creates. According to the model, all external forces are negative factors as they lower value of the company, so, the strategy of the traditional pipeline company should construct barriers to protect itself from these external forces. However, in the reality of the demand-side economy, external forces could be considered as positive factors that increase company's value. So, power of buyers and suppliers could be an asset for platforms. Moreover, while competitive forces and boundaries between them are quite clear in the supply-side economy, in the platform business reality those boundaries can shift quite rapidly. That is why, for platforms strategy, it is not necessary to build boundaries to protect itself from external forces, it is crucial to understand, when external forces decrease or increase company's value and how far boundaries between them can shift. For a better understanding let us look at examples of both accretive and depletive external forces. Buyers and suppliers usually create value for the platform. In the platform business model reality, most types of customer groups are interchangeable. Uber clients could be passengers one day and drivers next day. The same situation is for a tenant and a landlord for Airbnb clients. By such a role change, platform users increase its value even more. But role changes are not always positive for platforms. Sometimes buyers and suppliers could leave a platform by finding another place that meets their requirements more or even become such a place, starting their own platform and becoming a new competitor. This is the case for the Netflix. Being a content provider, the company uses telecom platforms in its own interests controlling interaction of customers with a content. Another example is the Samsung company, which was devices supplier for the Android platform, but then, decided to launch its own operational system by taking part of the customers with it. So, for platforms successful strategy, it is very important not only monitor its participants' activities, but also encourage activities on their platforms. That is why, those who have control over platforms, must wisely decide what buyers and suppliers could have access to the platform.

1.3 Factors driving business model adaptation to the multisided platform business model approach in retail industry.

Cambridge dictionary defines retail as “the activity of selling goods to the public, usually in shops”. (Cambridge Dictionary, 2016). Later definition of the same source implies that retail is “selling goods to the public in shops, on the internet, etc. ” With a dot.com boom and a spread of the Internet and its usage, it became possible to sell goods via the Internet. So, the term e-commerce (electronic commerce) has appeared, which means “the activity of buying or selling products or services electronically or over the Internet”. (Cambridge Dictionary, 2016). So, online retail platforms belongs to e-commerce industry.

Platforms are winners comparing to companies built on traditional linear business models due to the fact that there is no geographical restriction for platforms. Platform business models allow connecting buyers and sellers around the world, making both demand and offer wider and more competitive, thereby making a better match. But, besides this fact, there are also many other factors where platforms win and give traditional companies an incentive for a transition from the traditional business model to the platform. After analyzing the market, existing platforms and literature review on the topic, we identified the most significant factors. The majority of factors that make the company think about moving to the platform are inherent in all existing industries, but an impact from each of the factors manifests itself to a lesser or greater extent depending on the industry.

- Easier entrance to other fields

As we discussed before, platforms are winners in any industry which they enter due to such advantages as higher value creation for consumers through network effects and technological advantages and innovations - the constant change and implementation of which is one of the main components of platforms. Most entrepreneurs are mistaken in thinking that platforms will not enter their industry and that their industry cannot be adapted to the platform business. This has long been wrong. Moreover, platforms are not only already present in most of existing industries, but also easily move from one industry to another due to their advantages. Thus, companies like Amazon, starting with books selling, is currently one of the world's largest online retailers, selling not only all possible retail products, but also offering their users music and video content services, capturing more and more consumers and getting more and more profit.

...